End of training

README.md

@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.6944
-- Accuracy: 0.8205
+- Loss: 0.6981
+- Accuracy: 0.8194
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.9161290322580645,
-    "eval_loss": 0.23502571880817413,
-    "eval_runtime": 27.8973,
-    "eval_samples_per_second": 5.556,
-    "eval_steps_per_second": 1.398
+    "eval_accuracy": 0.8193548387096774,
+    "eval_loss": 0.6981304287910461,
+    "eval_runtime": 28.8048,
+    "eval_samples_per_second": 5.381,
+    "eval_steps_per_second": 1.354
 }

runs/May22_15-46-58_b08fa7a48520/events.out.tfevents.1716393129.b08fa7a48520.34.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e010ff645b41ba3d56da868c6fe62c1c19c151d94459fd20d163544e6e29f205
-size 411
+oid sha256:8356e43fd4bbfe107439d1e0c3c415b43b32e18da3aba6d845f2e27952bff7d7
+size 734

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.5,
-    "eval_accuracy": 0.9161290322580645,
-    "eval_loss": 0.23502571880817413,
-    "eval_runtime": 27.8973,
-    "eval_samples_per_second": 5.556,
-    "eval_steps_per_second": 1.398
+    "eval_accuracy": 0.8193548387096774,
+    "eval_loss": 0.6981304287910461,
+    "eval_runtime": 28.8048,
+    "eval_samples_per_second": 5.381,
+    "eval_steps_per_second": 1.354
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.9571428571428572,
+  "best_metric": 0.9428571428571428,
   "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-150",
   "epoch": 1.5,
   "eval_steps": 500,
@@ -10,152 +10,152 @@
   "log_history": [
     {
       "epoch": 0.07,
-      "grad_norm": 0.18143892288208008,
+      "grad_norm": 0.07965963333845139,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.0916,
+      "loss": 0.005,
       "step": 10
     },
     {
       "epoch": 0.13,
-      "grad_norm": 46.98843002319336,
+      "grad_norm": 0.08289032429456711,
       "learning_rate": 4.814814814814815e-05,
-      "loss": 0.2277,
+      "loss": 0.201,
       "step": 20
     },
     {
       "epoch": 0.2,
-      "grad_norm": 6.724725723266602,
+      "grad_norm": 0.2235719859600067,
       "learning_rate": 4.4444444444444447e-05,
-      "loss": 0.1532,
+      "loss": 0.5754,
       "step": 30
     },
     {
       "epoch": 0.27,
-      "grad_norm": 66.41813659667969,
+      "grad_norm": 0.0644197091460228,
       "learning_rate": 4.074074074074074e-05,
-      "loss": 0.5242,
+      "loss": 1.1323,
       "step": 40
     },
     {
       "epoch": 0.33,
-      "grad_norm": 0.5121557116508484,
+      "grad_norm": 42.82566452026367,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.4777,
+      "loss": 0.2046,
       "step": 50
     },
     {
       "epoch": 0.4,
-      "grad_norm": 44.47030258178711,
+      "grad_norm": 0.10422177612781525,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.7236,
+      "loss": 0.2749,
       "step": 60
     },
     {
       "epoch": 0.47,
-      "grad_norm": 32.01655960083008,
+      "grad_norm": 0.051237553358078,
       "learning_rate": 2.962962962962963e-05,
-      "loss": 0.1547,
+      "loss": 0.0433,
       "step": 70
     },
     {
       "epoch": 0.5,
-      "eval_accuracy": 0.9142857142857143,
-      "eval_loss": 0.35459208488464355,
-      "eval_runtime": 12.5328,
-      "eval_samples_per_second": 5.585,
-      "eval_steps_per_second": 1.436,
+      "eval_accuracy": 0.8857142857142857,
+      "eval_loss": 0.356876015663147,
+      "eval_runtime": 12.9523,
+      "eval_samples_per_second": 5.404,
+      "eval_steps_per_second": 1.39,
       "step": 75
     },
     {
       "epoch": 1.03,
-      "grad_norm": 0.12592756748199463,
+      "grad_norm": 0.14869919419288635,
       "learning_rate": 2.5925925925925925e-05,
-      "loss": 0.1536,
+      "loss": 0.1337,
       "step": 80
     },
     {
       "epoch": 1.1,
-      "grad_norm": 65.30023956298828,
+      "grad_norm": 0.21503913402557373,
       "learning_rate": 2.2222222222222223e-05,
-      "loss": 0.1538,
+      "loss": 0.061,
       "step": 90
     },
     {
       "epoch": 1.17,
-      "grad_norm": 31.011701583862305,
+      "grad_norm": 0.1664995551109314,
       "learning_rate": 1.8518518518518518e-05,
-      "loss": 0.2454,
+      "loss": 0.226,
       "step": 100
     },
     {
       "epoch": 1.23,
-      "grad_norm": 33.98771286010742,
+      "grad_norm": 0.13239744305610657,
       "learning_rate": 1.4814814814814815e-05,
-      "loss": 0.0889,
+      "loss": 0.1242,
       "step": 110
     },
     {
       "epoch": 1.3,
-      "grad_norm": 0.7245432138442993,
+      "grad_norm": 0.032610807567834854,
       "learning_rate": 1.1111111111111112e-05,
-      "loss": 0.0209,
+      "loss": 0.1647,
       "step": 120
     },
     {
       "epoch": 1.37,
-      "grad_norm": 0.1617116779088974,
+      "grad_norm": 0.5573199391365051,
       "learning_rate": 7.4074074074074075e-06,
-      "loss": 0.2224,
+      "loss": 0.1372,
       "step": 130
     },
     {
       "epoch": 1.43,
-      "grad_norm": 3.768871545791626,
+      "grad_norm": 0.3077225685119629,
       "learning_rate": 3.7037037037037037e-06,
-      "loss": 0.0144,
+      "loss": 0.0047,
       "step": 140
     },
     {
       "epoch": 1.5,
-      "grad_norm": 16.70306396484375,
+      "grad_norm": 0.22845041751861572,
       "learning_rate": 0.0,
-      "loss": 0.1088,
+      "loss": 0.0047,
       "step": 150
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.9571428571428572,
-      "eval_loss": 0.15032356977462769,
-      "eval_runtime": 12.6452,
-      "eval_samples_per_second": 5.536,
-      "eval_steps_per_second": 1.423,
+      "eval_accuracy": 0.9428571428571428,
+      "eval_loss": 0.3063296973705292,
+      "eval_runtime": 12.6147,
+      "eval_samples_per_second": 5.549,
+      "eval_steps_per_second": 1.427,
       "step": 150
     },
     {
       "epoch": 1.5,
       "step": 150,
       "total_flos": 7.47692094062592e+17,
-      "train_loss": 0.22406891534725826,
-      "train_runtime": 230.0097,
-      "train_samples_per_second": 2.609,
-      "train_steps_per_second": 0.652
+      "train_loss": 0.2195048264414072,
+      "train_runtime": 232.8628,
+      "train_samples_per_second": 2.577,
+      "train_steps_per_second": 0.644
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.23502568900585175,
-      "eval_runtime": 27.5068,
-      "eval_samples_per_second": 5.635,
-      "eval_steps_per_second": 1.418,
+      "eval_accuracy": 0.8205128205128205,
+      "eval_loss": 0.6944047212600708,
+      "eval_runtime": 28.6136,
+      "eval_samples_per_second": 5.452,
+      "eval_steps_per_second": 1.363,
       "step": 150
     },
     {
       "epoch": 1.5,
-      "eval_accuracy": 0.9161290322580645,
-      "eval_loss": 0.23502571880817413,
-      "eval_runtime": 27.8973,
-      "eval_samples_per_second": 5.556,
-      "eval_steps_per_second": 1.398,
+      "eval_accuracy": 0.8193548387096774,
+      "eval_loss": 0.6981304287910461,
+      "eval_runtime": 28.8048,
+      "eval_samples_per_second": 5.381,
+      "eval_steps_per_second": 1.354,
       "step": 150
     }
   ],